Probability Basics

Probability
A numerical measure of the likelihood of
occurrence of an event
Provides description of uncertainty

Experiments
A process that generates well-defined outcomes
At each repetition of experiment, one and only
one outcome will occur
Experiment
Flip a coin
Play a game
Roll a dice
Inspect a part

Experimental outcome
Head, tail (side 1, side
2) lose, tie
Win,
1, 2, 3, 4, 5, 6
Pass, fail

Sample Space
Sample space is the set of all
experimental outcomes
Flip a coin: S = {Head, Tail}
Roll a die: S = {1, 2, 3, 4, 5, 6}

Assigning Probabilities
Must adhere to two conditions:
The probability of an outcome:
0 P(Ei) 1 for all i
Ei = i th Experimental outcome

The sum of the probabilities of all outcomes:

P(Ei) = 1

Methods of assigning Probabilities

1. Classical Method
When it is reasonable to assume that all
outcomes are equally likely:
P(Ei)= 1/n, n= number of outcomes

Methods of assigning Probabilities

2. Relative Frequency Method
Relative frequency based on limited
repetition of the experiment
In a sample of 200 parts 10 are defective
P (defective) = 10/200 = 0.05

Methods of assigning Probabilities

3. Subjective Method
Used when the classical or the relative
frequency method are not possible, or as a
support for these methods
Probability is determined as the degree of
belief that an outcome will occur based on
available information and past experience
Assignment must ensure the two conditions
of assignment are applicable

Events and Their Probabilities

An event is a collection of outcomes
(sample points)
Roll a die experiment: define event A as
outcome is an even number

P (event) = sum of probabilities of sample

points
P (A) = P(2) + P (4) + P (6)
= 1/6 + 1/6 + 1/6 =

Complement of an Event

Union of Two Events

Intersection of Two Events

Addition Law

Mutually Exclusive Events

Events that are mutually exclusive do not
share any sample points in common
(probability of the intersection is zero)
Therefore:
P (A U B) = P (A) + P (B) P (A B)
Becomes
P (A U B) = P (A) + P (B)

Conditional Probability

Independent Events
Two events are independent if
P (B/A) = P (B) or
P (A/B) = P (A)
[The probability of one does not affect
the probability of the other]
Are mutually exclusive events
independent events?

Example
Suppose we have the following data for why
students consider joining a college (Question 13)
Reason for Joining
quality Cost Other Totals
Full time

421

393

76

890

Part time

400

593

46

1039

Totals

821

896

122

1929

Events
Event student is full time

Event student is part time

Event reason is Quality

Event reason is Cost

Event reason is Other

Joint Probability Table

Status

Reason for Joining

Quality Cost Other
(Q)
(C)
(O)
Full time
0.22 0.20 0.04
(F)
Part time 0.21 0.31 0.02
(T)
Totals
0.43 0.46 0.06
Joint probabilities

Marginal probabilities

Totals
0.46
0.54
1.00

Probabilities of Events

P (F) =
P (T) =
P (Q) =
P (C) =
P (O) =

Probabilities of Events

Student is FT ? P (Reason is
Quality)
Student is PT ? P (Reason is
Quality
Are T and Q independent?

Multiplication Law
Multiplication law is use to find the intersection
between two events
P (A B) = P (A/B) P(B)
P (A B) = P (B/A) P(A)
If A and B are independent then
P (A B) = P (A) P(B)

Bayes Theorem
Bayes theorem is used to update the
probabilities of events based on new
information (a sample, a test, etc.)

Example
A factory receives parts from two suppliers
A1 = Event part is from supplier 1 P (A1) = 0.65
A2 = Event part is from supplier 2 P (A2) = 0.35
G = Event part is good
B = Event part is bad
P (B/A1) = 0.0.02
P (G/A1) = 0.98
P (B/A2) = 0.0.02
P (G/A2) = 0.98
Suppose we select a bad part:
What is the probability that it came from supplier 1? P (A1/B)
What is the probability that it came from supplier 2? P (A2/B)

Probability Tree for Two-supplier Example

Bayes Theorem: Tabular Procedure

(1)

(2)
Prior
Event Prob

(3)
Cond
Prob

(4)
Joint
Prob

(5)
Posterior
Probability

P(Ai B)

P(Ai /B)

Ai

P(Ai) P(B/Ai)

A1

0.65 0.02

0.0130 0.0130/0.0305 = 0.4262

A2

0.35 0.05
1.00

0.0175 0.0175/0.0305 = 0.5738

P(B) = 0.0305
1.0000